Designing catalysts for nitrene transfer using early transition metals and redox-active ligands

Article abstract of DOI:10.1021/ic200911b

In this Forum Article, we discuss the use of redox-active pincer-type ligands to enable multielectron reactivity, specifically nitrene group transfer, at the electron-poor metals tantalum and zirconium. Two analogous ligand platforms, [ONO] and [NNN], are

Full text of DOI:10.1021/ic200911b

FORUM ARTICLE
pubs.acs.org/IC
Designing Catalysts for Nitrene Transfer Using Early Transition
Metals and Redox-Active Ligands
Alan F. Heyduk,* Ryan A. Zarkesh, and Andy I. Nguyen
Department of Chemistry, University of California, Irvine, California 92697-2025, United States
S Supporting Information
b
ABSTRACT:
In this Forum Article, we discuss the use of redox-active pincer-type ligands to enable multielectron reactivity, specifically nitrene
group transfer, at the electron-poor metals tantalum and zirconium. Two analogous ligand platforms, [ONO] and [NNN], are
discussed with a detailed examination of their similarities and differences and the structural and electronic constraints they impose
upon coordination to early transition metals. The two-electron redox capabilities of these ligands enable the transfer of organic
nitrenes to tantalum(V) and zirconium(IV) metal centers despite formal d0 electron counts. Under the correct conditions, the
resulting metal imido complexes can participate in further multielectron reactions such as imide reduction, nitrene coupling, or
formal nitrene transfer to an isocyanide.
are nitrene-exchange reactions (redox-neutral) rather than ni-
trene-transfer reactions (formal oxidations) owing to the inabil-
ity of the early transition metals to accommodate multielectron
changes in the oxidation state. For example, the reaction of an
alkene with a zirconium imido complex by a [2 + 2] cycloaddition
pathway forms a metalloazocyclobutane species, which can eliminate
an amine by protonolysis, regenerating the zirconium imide and
allowing for catalytic olefin hydroamination.17 While olefin
hydroamination by this pathway is not common, there have
been extensive reports of catalytic alkyne hydroaminations that
follow this route. In both olefin and alkyne hydroamination, the
overall transformation is a nonredox process as the product
amine is in the same oxidation state as the substrate and the metal
center stays in the zirconium(IV) oxidation state throughout the
process. In contrast, a two-electron redox reaction from the same
metalloazocyclobutane species would generate an aziridine by
nitrene transfer to the olefin (a formal oxidation of the double
bond). Such reactivity has not been observed for zirconium
imido complexes owing to the reluctance of zirconium(IV) to
undergo reductive elimination reactivity. In a few specific cases,
two-electron nitrene- or nitride-transfer chemistry has been
observed for group V metals, where access to both the 3+ and
5+ oxidation states is possible.18,19 To our knowledge, parallel
chemistry with group IV metals has not been reported, yet given
the bond-activation reactions summarized in Chart 1, accessible
redox reactivity that allows two-electron group transfer from
I. INTRODUCTION
Early-Transtion-Metal Imido Complexes. Early-transition-
metal imido complexes have attracted interest for the impressive
stoichiometric bond-activation reactions they can perform.1ꢀ4
Cyclopentadienylzirconium imido complexes provide an exam-
ple of these bond-activation reactions including [2 + 2] cycload-
ditions with alkynes, alkenes, imines, and isocyanides, as sum-
marized in Chart 1.5 CꢀH bond activation of aromatic and
aliphatic hydrocarbons is also possible with group IV imido
complexes. While Cp2ZrdNR complexes readily activate the
CꢀH bond of benzene to afford Cp2Zr(Ph)(NHR) derivatives,6
transient three-coordinate (silox)2M(dNR) complexes (M = Ti,
Zr; siloxꢀ = tri-tert-butyl siloxide) can activate both aromatic and
aliphatic CꢀH bonds, including those of methane, under mild
conditions.7,8 A key aspect of the exceptional reactivity displayed
by (silox)2M(dNR) is polarization of the metalꢀimido bond.
Juxtaposition of an electropositive zirconium or titanium metal
with an electronegative nitrene nitrogen leads to a highly
polarized MdNR bond, which, in turn, strongly polarizes the
substrate CꢀH bond for activation.9 In short, the electron-rich
imido moiety accepts a proton from the hydrocarbon, while the
electron-poor metal accepts the alkyl group. In the case of group
V imido complexes, less bond polarization leads to less reactive
complexes;10 however, these imido species still display impressive
bond-activation chemistry and continue to be of interest as potential
intermediates in homogeneous N2 reduction schemes.11ꢀ16
While the bond-activation chemistry of early-transition-metal
imido complexes has been exploited in the development of
some catalytic reactions, typically the overall transformations
Received: May 1, 2011
Published: July 20, 2011
r
2011 American Chemical Society
9849
dx.doi.org/10.1021/ic200911b Inorg. Chem. 2011, 50, 9849–9863
|

Inorganic Chemistry
FORUM ARTICLE
Chart 1
Chart 2
The implication of the simple model provided in Chart 2 is
that, for coordination complexes with strongly interacting metal
and ligand frontier orbitals, the metal and redox-active ligand can
work together to accommodate valence changes associated with
a redox reaction.29 Mapping this concept onto the zirconiumꢀ
imide bond-activation chemistry of Chart 1 suggests that redox-
active ligands may provide a mechanism for realizing formal
nitrene-transfer reactivity. Thus, a zirconium imido complex that
also contains a redox-active ligand might perform an initial
substrate activation step such as [2 + 2] cycloaddition or CꢀH
bond activation and follow it up with a two-electron, CꢀN bond-
forming reductive elimination to afford an aziridine or an amine,
respectively. The reductive elimination step would be made
possible because the redox-active ligand can accept two electrons,
circumventing the formation of a high-energy zirconium(II) species.
Adding a suitable nitrene-transfer reagent might then close a catalytic
cycle for olefin aziridination or CꢀH bond amination.
these metals would be attractive because it would open up a new
class of atom- and group-transfer catalysts.
Redox-Active Ligands. Redox-active ligands can exist in
multiple oxidation states when coordinated to a transition
metal.20ꢀ24 Catecholate is the prototypical redox-active ligand,
with three potential oxidation states: the dianionic catecholate,
the monoanionic open-shell radical semiquinonate, and the
neutral quinone. Upon coordination of catecholate-type ligands
to transition metals, even a cursory assignment of the metal
oxidation state becomes a challenging problem for both experi-
ment and theory.25,26 Chart 2 summarizes the electronic com-
plexity arising from the coordination of a redox-active ligand to a
redox-active metal. As shown across the top of Chart 2, resonance
structures can be drawn that change the distribution of electrons
between the redox-active ligand and the coordinated metal
without changing the total number of valence electrons con-
tained within the two fragments. In complexes with strong
metalꢀligand π interactions, these resonance structures help
to describe the complicated electronic interactions between
closely matched metal and ligand frontier orbitals within the
simple Lewis formalism. In some complexes, the different
electron distributions illustrated across the top of Chart 2 are
better described as electronic isomers (connected by equilibrium
arrows) because the electron distributions represent different
local minima on the potential energy surface.21,27,28
Moving vertically in Chart 2 represents redox transformations
within the metalꢀligand fragment that change the total number
of valence electrons contained within the unit. In classic transi-
tion-metal coordination chemistry, redox changes are considered
to occur mainly within the d-orbital manifold, so removing an
electron from a coordination complex results in an increase in the
metal oxidation state (metal-based oxidations). In complexes
with redox-active ligands, removal of an electron from the
complex can increase the oxidation level of the ligand (ligand-
based oxidation), leaving the metal oxidation state unchanged. As
illustrated in Chart 2, these two limiting processes give species
with the same number of valence electrons within the metalꢀ
ligand unit. With strong metalꢀligand π interactions, these
species again are best described as limiting resonance structures.
For the removal of a second electron from the metalꢀligand
fragment, it does not matter if the electron comes from the ligand
or from the metal because only a single resonance structure
and electron distribution is possible.
II. LIGAND DESIGN
In order to develop the type of metalꢀligand cooperative
redox reactivity hinted at in Chart 2, redox-active ligands are
needed with frontier orbitals that energetically match the early-
transition-metal d orbitals. The ligands also need to provide a
coordination environment that secures these reactive ions.
Earlier work from our laboratory on zirconium complexes of o-
phenylenediamine30 and o-aminophenol31,32 derivatives showed
that these catecholate-type ligands provide the desired electronic
matching of ligand and metal orbitals to achieve both oxidative
addition and reductive elimination reactivity (Scheme 1). It
became readily apparent that a common problem associated
with the use of bidentate redox-active ligands with d0 early-
transition-metal ions was the stability of the metal coordination
environment. Namely, while catecholate complexes of d0 metals
can show incredible formation constants and stability,33 upon
oxidation of amidophenolate ligands to the semiquinonate or
quinone oxidation states, ligand dissociation can be quite fast.
For example, the bis(amidophenolate) complex, [ap]2Zr(THF)2
(ap2ꢀ = 4,6-di-tert-butyl-2-tert-butylamidophenolate; THF = tetra-
hydrofuran) is stable to dissociation of the redox-active amidophe-
nolate ligand; however, upon two-electron oxidation to [isq]2ZrCl2
(isqꢀ =4,6-di-tert-butyl-2-tert-butyliminosemiquinonate),34 the com-
plex is unstable over long periods in solution, and ligand exchange
occurs to form [isq]4Zr complexes. When [ap]2Zr(THF)2 is
9850
dx.doi.org/10.1021/ic200911b |Inorg. Chem. 2011, 50, 9849–9863

Inorganic Chemistry
FORUM ARTICLE
Scheme 1
Chart 3
complexes were isolated as neutral species with complicated
electronic and magnetic properties. The three common oxida-
tion states of the [ONO] ligand are shown in Chart 3. In the fully
reduced or catecholate form, [ONOcat]3ꢀ, the ligand is closed-
shell and formally trianionic. Removal of one electron affords an
open-shell radical semiquinonate that is formally dianionic,
[ONOsq]2ꢀ. Removal of a second electron returns the ligand
to a closed-shell configuration as the monoanionic, quinonate
form of the ligand, [ONOq]ꢀ. The ligand typically coordinates to
metal ions in a meridional fashion, and this planar geometry
within the ligand allows for delocalization of the valence elec-
tronic changes over both six-membered rings. A synthetic route
for preparing and isolating the tridentate ligand in its “apo” form,
[ONOcat]H3, has been reported,41 and the [ONO] platform has
been used as a noninnocent ligand in the preparation of copper
oxidation catalysts for alcohol oxidation41 and as a supporting
ligand in the preparation of valence-tautomeric complexes.21
A redox-active pincer ligand analogous to the [ONO] plat-
form but with an all-nitrogen donor set is another attractive
target. The replacement of the oxygen donors of [ONO] with
nitrogen donors provides both different electronic properties and
tunable steric properties. An Ullman coupling between 2-nitro-
anisidine and 3-iodo-4-nitroanisole followed by Zn/HOAc reduc-
tion affords the backbone for an all-nitrogen redox-active ligand,
[NNNcat]H3.42 Treatment of the triammine with acetone and
NaBH3CN results in the installation of isopropyl groups onto the
primary amines via a reductive condensation. Trimethylsilyl43 and
amide44 derivatives also have been reported, but the redox proper-
ties of these derivatives have not been thoroughly developed. Three
oxidation states of the [NNN] ligand platform are shown alongside
those of the [ONO] ligand in Chart 3.
This manuscript provides an overview of our recent efforts to
utilize [ONO] and [NNN] redox-active, pincer-type ligands for
multielectron reactivity at electrophilic d0 metal centers. New
and previously reported42,45,46 results are discussed within the
context of developing nitrene-transfer reactivity for group IV and
V metal complexes. First, the electronic properties engendered
by the [ONO] and [NNN] ligand platforms are examined, with
consideration given to redox potentials, redox reactivity, and
spectroscopic interrogations of valence-electron distributions
between the redox-active ligand and coordinated metal ion.
Second, two-electron nitrene-group-transfer chemistry to
tantalum(V) and zirconium(VI) complexes of the [NNNcat]3ꢀ
ligand is discussed, highlighting the importance of the metal
coordination environment in the formation of monomeric and
oxidized by four electrons, iminoquinone is immediately ejected
from the metal coordination sphere.
To combat the dissociation of oxidized semiquinonate and
quinone ligands from the coordination spheres of d0 metal
complexes, we sought redox-active ligands with higher dentici-
ties. While the initial results with a tetradentate ligand derived by
condensing 2 equiv of 3,5-di-tert-butylcatechol with 1 equiv of
o-phenylenediamine were promising,35,36 the planar ligand topology
and tetraanionic charge (in its fully reduced form) made elabora-
tion of the ligand platform challenging. Furthermore, despite the
incorporation of three catecholate-type rings into the macro-
cyclic structure, the ligand could only be oxidized twice, with the
oxidation localized in the unprotected central phenylenediamide
fragment.
The so-called pincer-ligand family provides an alternative
topology for a redox-active ligand.37,38 A redox-active pincer
ligand can be prepared by stitching together two catechols
through condensation with ammonia. The in situ reaction of
3,5-di-tert-butylcatechol and ammonia in the presence of MX2
salts (e.g., M = Fe, Mn, Cu, and Zn) and air afforded M[ONO]2
complexes with octahedral geometries enforced by the meridio-
nal coordination of two tridentate amidobis(phenolates).39,40
The redox-activity of the ligands was evident because all of the
9851
dx.doi.org/10.1021/ic200911b |Inorg. Chem. 2011, 50, 9849–9863

Inorganic Chemistry
FORUM ARTICLE
Scheme 2
dimeric metal imide complexes and the subsequent reactivity of
the metal imide functionality. Third, the nitrene reactivity of
[ONO] complexes of tantalum(V) is discussed. These com-
plexes show a strong proclivity for mediating NꢀN bond
formation. The paper concludes with an outlook speculating
on the further development of multielectron and group-transfer
chemistry using early transition metals and redox-active ligands.
and 2203 cmꢀ1, respectively. Both of these vibrational frequen-
cies are blue-shifted relative to the CtN vibration for free tert-
butylisocyanide (2125 cmꢀ1), consistent with coordination of
the ligand to a d0 metal;47 however, the smaller blue shift of the
isocyanide vibrational frequency in [NNNcat]TaCl2(CNtBu) is
consistent with a less electron-poor tantalum center in the
[NNN] derivative. For further comparison, the CtN vibration
of the 16-electron tantalum(V) complex Cp*TaCl4(CNtBu) was
observed at 2241 cm ,
ꢀ1 48 while the CtN vibrations of the 18-
electron tantalum cations [(C5H4tBu)2TaV(H)2(CNtBu)]+ and
[Cp2TaIII(H2CdCHR)(CNtBu)]+ were observed at 2210 and
2164 cmꢀ1, respectively.49,50 These comparisons suggest that
[ONOcat]3ꢀ and [NNNcat]3ꢀ ligands are strongly electron-
donating toward the tantalum center, without formally reducing
the metal to the TaIII oxidation state. Consistent with this
hypothesis, attempts to make the analogous CO complexes,
[ONOcat]TaCl2(CO) and [NNNcat]TaCl2(CO), were unsuc-
cessful, providing further qualitative evidence that the tantalum
metal center is not in a highly reduced formal oxidation state
(i.e., TaIII).
III. BENCHMARKING REDOX-ACTIVE LIGAND
PROPERTIES
The [ONOcat]H3 and [NNNcat]H3 ligands are readily coordinated
to tantalum through the use of the mixed methyl chloride
tantalum synthon TaCl2Me3, as summarized in Scheme 2.42,46
The triprotic [NNNcat]H3 ligand is acidic enough to react with all
three methyl groups of TaCl2Me3 to release methane and afford
the five-coordinate complex [NNNcat]TaCl2 in good yield.
Similarly, the [NNNcat]H3 ligand reacts rapidly with TaCl2Me3
to release methane; however, in this case the isolated tantalum
complex is a six-coordinate Lewis basic adduct, [ONOcat]-
TaCl2L, with a coordinated diethyl ether, THF, or pyridine
ligand. The [ONO]Ta unit also may be accessed by taking
advantage of the different pKa's associated with the phenolic and
diphenylamine protons of [ONOcat]H3. Hence, deprotonation of
[ONOcat]H3 with 2 equiv of KH to generate the putative
[ONHO]K2, followed by the addition of TaCl2Me3, afforded
the dimethyl complex [ONOcat]TaMe2. The dimethyl complex
could be isolated as the five-coordinate, base-free species owing
to the strong electron-donating properties of the two methyl
ligands; however, in the presence of a strong Lewis base like
pyridine, the six-coordinate [ONOcat]TaMe2(py) adduct was
obtained.
The redox reactivity of [ONOcat]TaCl2L was benchmarked
using halogen oxidation reactions. The chlorine surrogate PhICl2
afforded [ONOsq]TaCl3 and [ONOq]TaCl4 upon one- and two-
electron oxidation of [ONOcat]TaCl2(OEt2), respectively, as
shown in Scheme 3.46 The one-electron oxidation product,
1
[ONOsq]TaCl3, is a paramagnetic, S = /2, complex, while the
two-electron oxidation product, [ONOq]TaCl4, is a diamagnetic
complex. The reaction progress is readily indicated by color
changes from red to blue to green as the [ONO] ligand is
oxidized from the catecholate to the semiquinonate to the
quinonate form.
Single-crystal X-ray diffraction studies provide unequivocal
data on the experimental metal and ligand oxidation states for
[ONOcat]TaCl2(OEt2), [ONOsq]TaCl3, and [ONOq]TaCl4.
Table 1 shows selected bond lengths for these three complexes.
Of note are the TaꢀClax bond distances (those perpendicular to
the [ONO]Ta plane), which fall in the 2.34ꢀ2.38 Å range across
the redox series. These distances are typical of chloride bond
distances in other tantalum(V) complexes. TaꢀCl bond dis-
tances within the same plane as the [ONO] ligand show greater
variability owing to steric crowding that occurs in the seven-
coordinate [ONOq]TaCl4 complex. Bond distances between the
tantalum center and the [ONO] ligand increase upon going from
[ONOcat]TaCl2(OEt2) to [ONOsq]TaCl3 to [ONOq]TaCl4,
consistent with oxidation of the [ONO] ligand across the series.
Isocyanide adducts of [ONOcat]TaCl2 and [NNNcat]TaCl2
were prepared so that vibrational spectroscopy could be used as a
probe of the electron-donating nature of the two redox-active
ligands. Treatment of [ONOcat]TaCl2(OEt2) with 1 equiv of
tert-butylisocyanide led to ligand substitution and formation of
the adduct [ONOcat]TaCl2(CNtBu). While [NNNcat]TaCl2
was isolated as the base-free five-coordinate complex, it reacted
quantitatively with the isocyanide to afford the analogous adduct
[NNNcat]TaCl2(CNtBu).42 In both complexes, the isocyanide
ligand coordinated trans to the nitrogen donor of the redox-
active ligand to afford C2v-symmetric species. The IR spectra of
[ONOcat]TaCl2(CNtBu) and [NNNcat]TaCl2(CNtBu) were
recorded as KBr pellets with CtN vibrations observed at 2225
9852
dx.doi.org/10.1021/ic200911b |Inorg. Chem. 2011, 50, 9849–9863

Inorganic Chemistry
FORUM ARTICLE
Scheme 3
Table 1. Comparison of Selected Bond Distances (Å) in the Ligand Oxidation Series [ONOcat]TaCl2(OEt2), [ONOsq]TaCl3, and
[ONOq]TaCl4
c
a This work. b See ref 46. c The equatorial plane is defined by the tantalum and the [ONO] ligand.
The larger isotropic hyperfine coupling constant, Aiso, of 31 G in
[NNNsq]TaCl3 is suggestive of greater delocalization of the
unpaired electron onto the tantalum center than in the [ONOsq]-
TaCl3 complex (Aiso = 20 G); however, both of these values are
smaller than those observed for organometallic tantalum(IV)
complexes (Aiso = 61ꢀ113 G)51 and for the tantalum(IV) co-
ordination complex TaCl4(PEt3)2 (Aiso = 211 G).52 Given prior
literature indicating that redox-active ligands derived from
o-phenylenediamine are easier to oxidize than those derived
from catechol or o-aminophenol,53 the inaccessibility of the two-
electron oxidation product, putative [NNNq]TaCl4, was surpris-
ing. To gain insight into whether this result was a consequence of
steric or electronic differences between the [ONO] and [NNN]
ligands, the two platforms were examined electrochemically.
Figure 1. X-band solution EPR spectra of (a) [ONOsq]TaCl3 and
To probe the ligand redox properties, the six-coordinate
(b) [NNNsq]TaCl3 collected at 298 K in THF. The spectra give g =
[ONOsq]TaCl3 and [NNNsq]TaCl3 complexes were investi-
1.979 for [ONOsq]TaCl3 and g = 1.897 for [NNNsq]TaCl3.
gated by cyclic voltammetry. Figure 2 shows the cyclic voltam-
metry data for both complexes collected in CH2Cl2 using a
platinum disk electrode and a 0.1 M [Bu4N][PF6] electro-
Congruous with this interpretation is the trend in the OꢀC and
NꢀC bond distances within the [ONO] ligand, which contract
as the ligand is oxidized.
lyte. All potentials are referenced to [Cp2Fe]+/0. [ONOsq]TaCl3
shows a reversible (ipa = ipc) one-electron reduction at ꢀ0.12 V
and a partially reversible (ipa < ipc) oxidation at +0.48 V. The
[NNNsq]TaCl3 analogue shows a reversible one-electron reduc-
tion at ꢀ0.32 V and a reversible one-electron oxidation at +0.15 V.
Two points about this data are worth emphasizing. First, the
electrochemical measurements clearly establish that the [NNN]
ligand platform is, in fact, easier to oxidize than the [ONO]
ligand platform. The oxidation from catecholate to semiquino-
nate is 200 mV easier for the [NNN] platform than it is for the
[ONO] platform; the oxidation from semiquinonate to quino-
nate is 300 mV easier for the [NNN] platform than it is for the
In contrast to the reactivity of [ONOcat]TaCl2(OEt2), the
reaction of [NNNcat]TaCl2 with PhICl2 affords only the single-
electron oxidation product [NNNsq]TaCl3 with no evidence for
the formation of [NNNq]TaCl4.42 The one-electron-oxidized
[NNNsq]TaCl3 product is a paramagnetic species that was
characterized by electron paramagnetic resonance (EPR) spec-
troscopy. Figure 1 shows the X-band solution EPR spectra of
both [ONOsq]TaCl3 and [NNNsq]TaCl3 at 298 K. Both spectra
show isotropic room-temperature signals consistent with S = 1/2
spin systems and hyperfine coupling to I = 7/2 tantalum nuclei.
9853
dx.doi.org/10.1021/ic200911b |Inorg. Chem. 2011, 50, 9849–9863

Inorganic Chemistry
FORUM ARTICLE
Scheme 4
Figure 2. Cyclic voltammograms of (a) [ONOsq]TaCl3 and (b)
[NNNsq]TaCl3 in CH2Cl2 with 0.1 M [Bu4N][PF6] at 200 mV sꢀ1
.
The stars (f) denote the potentials of the bulk tantalum complex.
Potentials are referenced to [Cp2Fe]+/0
.
[ONO] platform. Second, the semiquinone-to-quinone oxida-
tion of the [NNN]Ta species is more reversible than the
corresponding oxidation of the [ONO]Ta species. We attribute
this difference in reversibility to differences in the steric con-
straints imposed by the [ONO] and [NNN] platforms. Oxida-
+
tion of [ONOsq]TaCl3 to [ONOq]TaCl3 generates an
suggest that both ligand platforms can access the catecholate,
semiquinonate, and quinonate oxidation states, the potentials for
the [NNN] redox-active ligand are 200ꢀ300 mV lower than
those for the [ONO] redox-active ligand. Second, despite this
clear evidence that the [NNN] ligand is easier to oxidize than the
[ONO] derivative, chemical oxidations using chlorine surrogates
indicate that sterics play a big part in stabilizing the oxidation
products. Given these results, it seemed plausible that two-
electron reactivity would be possible for the [NNN]Ta platform,
provided that seven-coordinate species could be avoided.
electrophilic and reactive cation. As illustrated by the crystal
structure of [ONOq]TaCl4, the [ONO] ligand platform can
accommodate a seven-coordinate geometry at tantalum. In the
electrochemical experiment, the [ONOq]TaCl3 cation likely
+
adds whatever ligand it can find in solution (for example, by
binding a [PF6]ꢀ anion),54ꢀ56 leading to irreversibility in the
[ONOsq]TaCl3 to [ONOq]TaCl3+ redox process (EC reaction).
In contrast, the increased steric constraints imposed by the iPr
groups of the [NNN] ligand disfavor seven-coordinate geome-
tries at tantalum and prevent the coordination of Lewis bases to
+
the electrophilic tantalum center of [NNNq]TaCl3 ; therefore,
on the electrochemical time scale, the [NNNsq]TaCl3 to
IV. NITRENE-TRANSFER REACTIVITY OF [NNNcat]TaCl2
[NNNq]TaCl3+ redox process is reversible.
Combining the electrochemical results presented in Figure 2
with the halogen reactivity of [ONOcat]TaCl2L and [NNNcat]-
TaCl2 provides insight into the factors governing the reactivity of
these two platforms. In the case of the [ONO] ligand, the two-
electron oxidation product, [ONOq]TaCl4, is an isolable com-
plex thanks to the stabilization provided by coordination of the
fourth halide to the tantalum center. In the case of the [NNN]
ligand, the first chlorine atom addition to [NNNcat]TaCl2 gives
six-coordinate [NNNsq]TaCl3. The addition of the second
chlorine atom should (and probably does) oxidize the [NNN]
ligand up to the quinone oxidation state; however, the N(iPr)
groups block access to a seven-coordinate structure and, hence,
coordination of the fourth chloride ligand. While the putative
On the basis of the identification of two reversible one-
electron redox processes for the [NNN]Ta platform, we sought
to develop two-electron group-transfer reactivity. According to
Scheme 4, the treatment of [NNNcat]TaCl2 with aryl azides
resulted in liberation of N2 and concomitant formation of
the tantalum imide complex [NNNq]TaCl2(dNAr) (Ar = Ph,
p-tolyl, 4-C6H4tBu, 3,5-C6H3Me2).42 Azides with more sterically
demanding aryl substituents (i.e., 2,6-C6H3iPr2) did not react
with [NNNcat]TaCl2 nor did alkyl azides or trimethylsilyl azide.
Similarly, the Lewis-base adduct [NNNcat]TaCl2(py) did not
react with aryl azides, highlighting the importance of an open
coordination site at the metal center. Figure 3 shows the structure
of [NNNq]TaCl2(dNPh) along with pertinent bond distances
within the metal coordination sphere and the [NNN] ligand. The
TaꢀN and TaꢀCl bond distances are typical for a tantalum(V)
complex. The bond distance for the tantalum imide is consistent
with the triple-bond character between a d0 tantalum center and
an imide ligand acting as a six-electron donor.12,57ꢀ59 Consistent
with this contention is the TaꢀNꢀC bond angle within the
imide, which is close to linear at 167.4(4)°. Within the [NNN]
ligand, the bond distances clearly indicate the quinone oxidation
state, with the pattern of CꢀC bond distances consistent with
cyclohexadiene rings.60
+
[NNNq]TaCl3 cation is stable on the electrochemical time
scale, it may not be stable enough to allow for isolation (repeated
attempts to isolate this species with a variety of noncoordinating
anions failed); a possible decomposition pathway may involve
hydrogen-atom abstraction from the methine carbon of a
N(iPr) group.
Two key observations can be made regarding the reactivity
patterns of the [ONO]Ta and [NNN]Ta platforms: the first an
electronic effect and the second a steric effect. First, vibrational
spectroscopy and electrochemical analyses both suggest that the
[NNN] redox-active ligand provides a more reducing environ-
ment for the coordinated metal than the oxygen derivative
[ONO] redox-active ligand. While the electrochemical data
The imide in [NNNq]TaCl2(dNAr) can be reduced with
hydrazine, but attempts to react it with other organic substrates
have been unsuccessful. [NNNq]TaCl2(dNPh) reacts cleanly
9854
dx.doi.org/10.1021/ic200911b |Inorg. Chem. 2011, 50, 9849–9863

Inorganic Chemistry
FORUM ARTICLE
Figure 5. ORTEP diagram of [NN(Me)N]TaMe(dN-p-tolyl). Ther-
mal ellipsoids are shown at 50% probability, and hydrogen atoms are
omitted for clarity.
hydrogen-atom donors such as 9,10-dihydroanthracene or 1,4-
cyclohexadiene did not result in reduction of the imido complex,
nor did any [NNNq]TaCl2(dNAr) react with H2 or the CꢀH
bonds of benzene, toluene, or other alkanes. Attempts to transfer
the nitrene to unsaturated organic substrates such as R-olefins
and alkynes were unsuccessful. These results suggest that while
[NNNq]TaCl2(dNAr) is susceptible to proton-coupled elec-
tron-transfer reactivity, it is not capable of two-electron nitrene-
transfer chemistry analogous to mid- and late-transition-metal
imido complexes.
Figure 3. ORTEP diagram of [NNNq]TaCl2(dNPh). Thermal ellip-
soids are shown at 50% probability, and hydrogen atoms are omitted for
clarity. Bond distances are given in angstroms to the hundredths decimal
place. Data were taken from a high-resolution structure in which the
errors in the CꢀO,N,C distances were <0.01 Å.42
Attempts to generate organometallic tantalum imide com-
plexes resulted in alkylation of the [NNN] ligand.63 The addition
of 2 equiv of MeLi to solutions of [NNNq]TaCl2(dNAr)
complexes led to N-methylation of the redox-active ligand and
formation of [NN(Me)N]TaMe(dNAr). Single crystals of one
derivative, [NN(Me)N]TaMe(dN-p-tolyl), were obtained and
provided the molecular structure shown in Figure 5. Methylation
of the ligand is readily apparent in the structure and results in the
[NN(Me)N] ligand adopting more of a facial coordination mode
(NꢀTaꢀN 114°). The TaꢀN distance to the methylated
nitrogen is long at 2.38 Å, consistent with this nitrogen acting
as a neutral amine donor. The TaꢀN distance to the imide ligand
remains short at 1.80 Å, and the TaꢀNꢀC bond angle is opened
1
further (173°) relative to [NNNq]TaCl2(dNPh). The H and
13C NMR spectra of [NN(Me)N]TaMe(dN-p-tolyl) are en-
tirely consistent with the X-ray structure. The resonance for the
[NN(Me)N] methyl group appears as a singlet at 3.03 ppm in
the 1H NMR spectrum and the tantalum-bound methyl appears
at 0.83 ppm. Reactions between [NNNq]TaCl2(dNAr) and
1 equiv of MeLi did not proceed as cleanly as those with 2 equiv
of MeLi; however, according to NMR spectroscopy, the primary
product of those reactions is the ligand-methylated species,
[NN(Me)N]TaCl(dNAr). This result suggests that the initial
reaction between [NNNq]TaCl2(dNAr) and MeLi is nucleo-
philic attack of Meꢀ at the quinonate ligand, a contention that
is supported by the observed geometry in the [NN(Me)N]-
TaMe(dN-p-tolyl) product.
We surmised that a potential problem with utilizing
[NNNq]TaCl2(dNAr) as a platform for nitrene transfer to
substrates was coordinative saturation at the tantalum metal
center. [NNNq]TaCl2(dNAr) is a six-coordinate, 16-electron
complex. The N(iPr) groups of the [NNN] ligand effectively
block access to the metal center, preventing coordination of a
substrate that can accept the nitrene. In order to examine the
reactivity of the imido fragment with reductants other than
hydrazines, it would be necessary to open a coordination site
at the metal center. Attempts to abstract a chloride ligand from
Figure 4. Changes in the electronic absorption spectrum during the
reaction of [NNNq]TaCl2(dNPh) with excess 1,2-diphenylhydrazine
in THF. Spectra were recorded in 30 min intervals at 35 °C.
with 1,2-diphenylhydrazine to afford aniline and azobenze ac-
cording to Scheme 4. When the reaction was carried out in THF
at room temperature, there was no evidence for the buildup of an
intermediate species such as [NNNsq]TaCl2(NHPh), which
would be expected to have a distinctive absorption in the visible
portion of the spectrum. Instead, the spectrum for the imido
complex [NNNq]TaCl2(dNPh) cleanly decays to the spectrum
of the fully reduced [NNNcat]TaCl2 (Figure 4). Reactions
carried out using [NNNq]TaCl2(dN-4-C6H4tBu) and 1,2-di-
phenylhydrazine yielded the exclusive formation of p-tert-butyl-
aniline and azobenzene, indicating an hydrogen-atom-transfer
or a proton-coupled electron-transfer pathway61,62 rather than
a metathesis pathway involving cleavage of the NꢀN bond
of the hydrazine reductant. Reactions with other potential
9855
dx.doi.org/10.1021/ic200911b |Inorg. Chem. 2011, 50, 9849–9863

Inorganic Chemistry
FORUM ARTICLE
the central nitrogen of the [NNN] ligand. As with the analogous
tantalum complex, [NNNcat]TaCl2(CNtBu), the zirconium de-
rivative [NNNcat]ZrCl(CNtBu)2 shows the CtN stretch to
higher frequency (2201 cmꢀ1) than the CtN stretch in free
isocyanide (2125 cmꢀ1), consistent with an electron-poor metal
center.
Aryl azides react with [NNNcat]ZrCl(THF)2, transferring the
nitrene to the metal center and yielding a dimeric zirconium
imide complex. Conversion of [NNNcat]ZrCl(THF)2 to the
oxidized imido dimer {[NNNq]ZrCl(μ-NAr)}2 (Ar = Ph,
4-C6H4tBu, p-tolyl, 4-C6H4CF3, 2,4,6-C6H2Me3) was readily
monitored by optical spectroscopy as the orange starting material
gives way to an olive-green product with an intense absorption
band at ∼430 nm. While the complexes show overall C2v
symmetry by NMR spectroscopy, two inequivalent imido ligands
are evident. X-ray-quality crystals of the 4-tert-butylphenylimido
derivative were obtained, and the structure is shown as an
ORTEP diagram in Figure 6. The zirconium dimer forms with
the chloride ligands occupying syn positions with respect to the
Zr Zr vector, an arrangement that forces the two [NNN]
3 3 3
ligands to sandwich one bridging imide ligand. As with the
monomeric tantalum analogue, the bond lengths within the
[NNN] ligand indicate that both ligands are in the quinonate
oxidation state.
Despite the chemically distinct environments of the two
bridging imido ligands in the {[NNNq]ZrCl(μ-NAr)}2 com-
plexes, they proved to be unreactive. Heating mixtures of
{[NNNq]ZrCl(μ-NPh)}2 and {[NNNq]ZrCl(μ-N-p-tolyl)}2
for days in benzene did not lead to crossover formation of the
putative mixed imido complex {[NNNq]ZrCl}2(μ-NPh)(μ-N-
p-tolyl). Similarly, olefins, alkynes, phosphines, sulfides, and
isocyanides failed to react with the imido dimers under thermal
conditions. Presumably, the aryl azides react with [NNNcat]-
ZrCl(THF)2 to form a terminal imido complex initially, [NNNq]-
ZrCl(dNAr)(THF), which then loses THF and dimerizes to
form the observed product. The unreactive nature of this dimeric
complex under thermal conditions indicates that formation of the
bridging imide core is not reversible under thermally accessible
conditions; thus, any transfer of the nitrene from the metal center
to a substrate must occur prior to dimerization of the putative
terminal imide intermediate.
Figure 6. ORTEP diagram of {[NNNq]ZrCl(μ2-N-4-C6H4tBu)}2.45
Thermal ellipsoids are shown at 50% probability. Hydrogen atoms and a
diethyl ether solvent molecule are omitted for clarity.
[NNNq]TaCl2(dNAr) with various boranes and alumanes did
not afford the desired five-coordinate [NNNq]TaCl(dNAr)+
cation. As an alternative method to generating a five-coordinate
imido complex, analogous [NNN] complexes of zirconium were
prepared.
The bis(isocyanide) complex, [NNNcat]ZrCl(CNtBu)2, of-
fered a way to intercept a putative terminal zirconium imide
intermediate. The treatment of [NNNcat]ZrCl(CNtBu)2 with
1 equiv of 4-tBuC6H4N3 resulted in liberation of 1 equiv of N2 gas
and formation of the carbodiimide 4-tBuC6H4NdCdNtBu.64,65 The
remaining zirconium complex was characterized by the addi-
tion of pyridine to the solution upon completion of the reaction
and gave quantitative recovery of [NNNcat]ZrCl(py)2. In the
presence of excess aryl azide and excess tBuNC, up to 2 equiv of
carbodiimide was formed per zirconium complex. When the alkyl
azides tBuN3 or AdN3 were used in excess with excess tBuNC,
the catalytic formation of carbodiimide was observed corre-
sponding to approximately 10 turnovers of the zirconium com-
plex. Attempts to run the nitrene-transfer reaction with lower
catalyst loadings were unsuccessful because the reaction is
impeded both by isocyanide and by the carbodiimide product.
Replacement of isocyanide with CO resulted in no nitrene
transfer to form isocyanate. A simplified mechanism for the
nitrene-transfer reaction that takes into account these observa-
tions is shown in Scheme 5. The inverse dependence of the
reaction rate on the isocyanide concentration suggests that the
V. CATALYTIC NITRENE TRANSFER BY
[NNNcat]ZrCl(CNtBu)2
45
Coordination of the [NNN] ligand platform to zirconium
affords a six-coordination, pseudooctahedral complex, [NNNcat]-
ZrCl(THF)2. Metalation was readily achieved by treating ZrCl4-
(THF)2 with the deprotonated ligand, [NNNcat]Li3(OEt2)n (formed
in situ), in diethyl ether. While the product, [NNNcat]ZrCl(THF)2,
is a saturated six-coordinate complex, two of the coordination
sites are taken up by substitutionally labile THF ligands. Indeed,
the 1H NMR spectrum of [NNNcat]ZrCl(THF)2 in the presence
of free THF shows only a single set of THF resonances at room
temperature, consistent with rapid exchange between free and
coordinated THF on the NMR time scale. The THF ligands are
readily displaced by more strongly donating ligands, and as such
the addition of 2 equiv of pyridine or tert-butylisocyanide to
[NNNcat]ZrCl(THF)2 resulted in the quantitative formation of
[NNNcat]ZrCl(py)2 or [NNNcat]ZrCl(CNtBu)2, respectively.
In all three complexes, the NMR spectra show C2v-symmetric
species, suggesting that the chloride ligand is coordinated trans to
9856
dx.doi.org/10.1021/ic200911b |Inorg. Chem. 2011, 50, 9849–9863

Inorganic Chemistry
FORUM ARTICLE
Scheme 5
first step of the reaction is dissociation of an isocyanide to form a
[NNNcat]ZrCl(CNtBu) species. This five-coordinate species can
be oxidized by the organic azide to form the putative monomeric
imide intermediate, [NNNq]ZrCl(dNR)(CNtBu), in the rate-
determining step. It is established that early-transtion-metal
imido complexes with isocyanide ligands undergo nonredox,
migratory insertion reactions to form η2-carbodiimide species.5
In this case, the migratory insertion step disposes the complex for
a second CꢀN bond-forming reaction, reductive elimination,
which forms the CꢀN π bond and reduces the [NNNq]ꢀ ligand
back to the [NNNcat]3ꢀ form. The final step to close the catalytic
cycle is displacement of the carbodiimide product from the metal
coordination sphere by an isocyanide. Because the final step is a
ligand-exchange process, product buildup impedes the reaction
at high conversion.
plausibly [ONOq]TaCl2(dNR), along the reaction pathway.
In an attempt to study this putative imido intermediate, alter-
native routes to transfer a nitrene to [ONOcat]TaCl2(OEt2) were
investigated.
One potential route to the formation of metal imido com-
plexes is the homolytic cleavage of organic diazenes.15,66 The
addition of azobenzene to solutions of [ONOcat]TaCl2(THF) or
[ONOcat]TaCl2(py) resulted in no reaction; however,
when azobenzene was added to a benzene solution of
[ONOcat]TaCl2(OEt2), the red color immediately changed to
green. Solutions of the green product were paramagnetic with
μeff = 1.37 μB (Evans’ method67 at 298 K); upon cooling to 188 K,
the magnetic moment increased to 2.18 μB. The addition of THF
or pyridine to these green solutions resulted in an immediate
and quantitative conversion to [ONOcat]TaCl2(THF) or
[ONOcat]TaCl2(py), respectively, with the release of free azo-
benzene. When [ONOcat]TaCl2(OEt2) was reacted with a
mixture of PhNdNPh and (p-tolyl)NdN(p-tolyl) for several
hours at room temperature and then quenched with pyridine,
GC/MS analysis of the organic products showed no evidence for
scrambling of the diazenes, indicating that complete scission of
the diazene NꢀN bond did not occur. Titration experiments
were conducted to examine formation of the paramagnetic green
product, and representative UVꢀvis spectroscopic data are
shown in Figure 7. Upon the addition of azobenzene to solutions
of [ONOcat]TaCl2(OEt2), new bands grow in at 590 and
950 nm, consistent with the semiquinonate form of the
[ONO] ligand. Monitoring either band by Job’s method68
(Figure 7, inset) revealed a ratio of two tantalum complexes to
one azobenzene in the green product. The data are consistent
with assignment of the green species as a tantalum dimer,
{[ONO]TaCl2}2(μ2-N2Ph2), shown in Scheme 7. Coordination
of the azobenzene between two [ONOcat]TaCl2 fragments
should raise the azobenzene reduction potentials69 within the
range of [ONO] ligands, resulting in the transfer of one or two
electrons to the azobezene and with concomitant oxidation of
one or both [ONO] ligands to the semiquinonate level. The
temperature dependence of μeff is consistent with the proposed
VI. NITRENE CHEMISTRY OF THE [ONO]Ta PLATFORM
The two-electron reactivity of the [ONO]Ta platform, dis-
cussed in section III, prompted further investigations into the
reactivity of that system. Accordingly, oxidation of the imido
dimer complex {[ONOcat]Ta(py)(μ-NPh)}2 with 2 equiv of
PhICl2 resulted in the formation of azobenzene and 2 equiv of
[ONOcat]TaCl2(py), an overall four-electron reductive elimina-
tion of azobenzene (Scheme 6).46 Reactions carried out with
substoichiometric quantities of oxidant resulted in a portion of
the starting materials being carried through to the products with
no evidence for the formation of a partially oxidized intermediate.
An interesting aspect of the oxidation of {[ONOcat]Ta(py)
(μ-NPh)}2 by PhICl2 was revealed in crossover experiments.
The unoxidized dimer {[ONOcat]Ta(py)(μ-NPh)}2 is robust in
solution, showing no evidence for dissociation into terminal
monomeric tantalum imido complexes. In contrast, when a
1:1 mixture of {[ONOcat]Ta(py)(μ-NPh)}2:{[ONOcat]Ta(py)
(μ-N-p-tolyl)}2 was oxidized with a stoichiometric quantity of
PhICl2, the product azobenzenes were a statistical distribution
of the derivatives PhNdNPh, (p-tolyl)NdN(p-tolyl), and
PhNdN(p-tolyl), implicating a terminal imido intermediate,
9857
dx.doi.org/10.1021/ic200911b |Inorg. Chem. 2011, 50, 9849–9863

Inorganic Chemistry
FORUM ARTICLE
Scheme 6
prepare the putative terminal imide of the [ONO]Ta platform
using the same strategy. Again, aryl azides did not react with
[ONOcat]TaCl2(THF) or [ONOcat]TaCl2(py), presumably
because the organoazide cannot displace the THF and pyridine
ligands from the tantalum center, even at elevated tempera-
tures. In contrast, the addition of PhN3 to the ether adduct,
[ONOcat]TaCl2(OEt2), at 55 °C in benzene resulted in the
liberation of 1 equiv of nitrogen gas (quantified by a Toepler
pump) and the formation of a green solution, which returned
[ONOcat]TaCl2(py) in approximately 60% yield upon the addi-
tion of pyridine. Azobenzene was the organic product, formed in
quantitative yield according to GC/MS analysis. The addition of
excess PhN3 to solutions of [ONOcat]TaCl2(OEt2) resulted in
the release of approximately 10 equiv of N2 and the formation of
5 equiv of azobenzene; however, upon the addition of pyridine,
only a trace of [ONOcat]TaCl2(py) could be isolated, with the
bulk of the tantalum-containing product being [NNNsq]TaCl3
and an unidentified paramagnetic product. The discovery of
[ONOsq]TaCl3 in reactions with excess organic azides suggested
a decomposition pathway involving the expulsion or transfer of
chlorine atoms (Cl•). To circumvent this side reaction, more
strongly binding alkoxides were used to replace the chloride
coligands.
Figure 7. Changes in the electronic absorption spectrum upon titration
of [ONOcat]TaCl2(OEt2) with azobenzene in benzene at 22 °C. Inset:
Job’s plot for the reaction of [ONOcat]TaCl2(OEt2) with azobenzene.
equilibrium formation of {[ONO]TaCl2}2(μ2-N2Ph2) as a ground-
state S = 1 species; however, we have been unable to isolate pure
samples of the complex in the solid state to conduct the requisite
magnetic measurements to verify this hypothesis.
Given the facile transfer of nitrene from organoazides to
[NNNcat]TaCl2 and [NNNcat]ZrCl(L)2, we attempted to
Scheme 7
9858
dx.doi.org/10.1021/ic200911b |Inorg. Chem. 2011, 50, 9849–9863

Inorganic Chemistry
FORUM ARTICLE
Scheme 8
catalyze the nitrene coupling reaction, consuming 10 equiv of
phenyl azide to produce 5 equiv of azobenzene over the span of
7 days at 55 °C. After complete consumption of the azide,
[ONOcat]Ta(OtBu)2 was recovered quantitatively with no evi-
dence for the formation of paramagnetic side products. Evidence
for the intermediacy of the putative imido complex, [ONOq]-
Ta(OtBu)2(dNAr), was obtained when [ONOcat]Ta(OtBu)2
was reacted with 1 equiv of (p-tolyl)N3 in the presence of excess
1,2-diphenylhydrazine (Scheme 8). Under these conditions, no
(p-tolyl)NdN(p-tolyl) or (p-tolyl)NdNPh was observed; in-
stead, p-tolylnitrene was converted exclusively to p-toluidine,
exactly analogous to the observed hydrazine reduction of
[NNNq]TaCl2(dNPh).
Figure 8. ORTEP diagram of [ONOcat]Ta(OtBu)2. Thermal ellipsoids
are shown at 50% probability. Hydrogen atoms and a pentane solvent
molecule are omitted for clarity. Bond distances are given in angstroms
to the hundredths decimal place. Data were taken from a high-resolution
structure in which the errors in CꢀO,N,C distances were <0.01 Å; see
the Supporting Information.
Monomeric [ONOcat]Ta(OtBu)2 was accessible by a one-pot,
two-step procedure. Attempts to prepare [ONOcat]Ta(OtBu)2
by the metathesis reaction of [ONOcat]TaCl2(OEt2) with LiOt-
Bu led to an intractable product mixture. Instead, [ONOcat]-
Ta(OtBu)2 was obtained by the conversion of [ONOcat]-
TaCl2(OEt2) to [ONOcat]TaMe2 with MeLi followed by the
dropwise addition of tBuOH to give the desired alkoxide as a
pale-yellow solid. The 1H NMR spectrum of [ONOcat]-
Ta(OtBu)2 showed three singlets in the tert-butyl region in a
1:1:1 ratio (1.68, 1.45, and 1.18 ppm). NMR spectroscopy
showed no evidence for the binding of solvent to form a six-
coordinate complex. An X-ray diffraction study, undertaken on a
single crystal of [ONOcat]Ta(OtBu)2, confirmed a five-coordi-
nate geometry around the tantalum center. Figure 8 shows the
ORTEP of [ONOcat]Ta(OtBu)2 along with pertinent bond
distances within the metal coordination sphere and the [ONO]
ligand.
VII. LESSONS GOING FORWARD
As with the development of any metal coordination complex
for small-molecule reactivity, both electronic and steric factors
have an important part to play. Within the results discussed in
this work, both [ONO] and [NNN] ligands provide an appro-
priate redox reservoir to enable multielectron reactivity at
tantalum(V) and zirconium(IV) metal centers. While experi-
mental data indicate that these ligands prefentially hold onto the
redox-active pair of valence electrons, keeping the metals in their
maximum oxidation states, both chlorine and nitrene chemistry
indicates that the electron pair is accessible for two-electron
oxidants (2Cl• or :NR) to add at the metal center. From this
perspective, [ONO] and [NNN] ligands can be viewed as
cofactors that enable redox reactions at the metal center. The
differences in redox potentials observed for [NNN] and [ONO]
do manifest differences in reactivity, the most obvious
being in the coupling of arylnitrenes to form aryldiazenes.
[ONOcat]TaX2(OEt2) (X = Cl or OtBu) reacts with aryl azide to
afford the aryldiazene in quantitative yield. Evidence suggests
that this reaction proceeds through the initial formation of
a terminal tantalum imide, [ONOq]TaX2(dNAr), which di-
merizes to {[ONOq]TaX2(μ2-NAr)}2 to eliminate diazene.
The elimination of diazene from the putative imido dimer is
an overall four-electron reductive elimination that reduces
two [ONOq]ꢀ to two [ONOcat]3ꢀ. For the [NNN] ligand set,
[NNNcat]ZrCl(THF)2 reacts with aryl azides along an analogous
path to form an imide dimer with two quinonate ligands,
{[NNNq]ZrCl(μ2-NAr)}2. This imido dimer is stable indefi-
nitely and does not extrude the aryldiazene, even at elevated
The dialkoxide complex [ONOcat]Ta(OtBu)2 is a competent
catalyst for azobenzene formation. The reaction of [ONOcat]-
Ta(OtBu)2 with PhN3 proceeded at 55 °C to form PhNdNPh
(Scheme 8), albeit at a reduced rate relative to reactions
employing [ONOcat]TaCl2(OEt2). Whereas reactions between
[ONOcat]TaCl2(OEt2) and 1 equiv of PhN3 were complete
within 1 h at 55 °C, those between [ONOcat]Ta(OtBu)2 and
PhN3 took 24 h. The longer reaction times are attributed to the
sterically demanding tert-butyl groups of the alkoxide ligands,
which hinder both initial coordination of the azide to the metal
center and subsequent expulsion of N2 to form the tantalum
imide.70 Unlike the nitrene coupling reactions employing
[ONOcat]TaCl2(OEt2), in the presence of excess PhN3,
[ONOcat]Ta(OtBu)2 does not decompose and continues to
9859
dx.doi.org/10.1021/ic200911b |Inorg. Chem. 2011, 50, 9849–9863

Inorganic Chemistry
FORUM ARTICLE
temperatures, despite having two [NNNq]ꢀ ligands to accept the
four electrons from diazene reductive elimination. It may be the
case that the [ONOq]ꢀ ligand can pull electrons away from
the bridging imide ligands and force diazene elimination, whereas
the [NNNq]ꢀ ligand is not oxidizing enough to promote the
reaction. Moving forward, catalyzing nitrene-transfer reactions to
synthetically interesting substrates will require that the redox
potentials of the ligand are balanced carefully with both the
nitrene donor and the nitrene acceptor in order for the metal
complex to act as a nitrene-transfer catalyst.
notation δ in parts per million. IR spectra were recorded as KBr pellets
with a Perkin-Elmer Spectrum One FTIR spectrophotometer. Elec-
tronic absorption spectra were recorded as solutions in 1 cm quartz cuvettes
using either a Perkin-Elmer Lambda 800 UVꢀvis spectrophotometer or
a Perkin-Elmer Lambda 800 UVꢀvis spectrophotometer equipped with
a near-IR detecor. EPR spectra were collected on a Bruker EMX X-band
spectrometer equipped with an ER041XG microwave bridge. Spectra for
EPR samples were collected using the following spectrometer settings:
attenuation = 20 dB, microwave power = 2.017 mW, frequency =
9.64 GHz; sweep width = 300 G, modulation amplitude = 9.02 G, gain =
2.00 ꢁ 103, conversion time = 10.24 ms, time constant = 81.92 ms, and
resolution = 2048 points. Elemental analyses were collected on a Perkin-
Elmer 2400 series II CHNS/O analyzer. GC/MS data were collected
using a Finnigan Trace mass spectrometer with a 30 m DB-5 column.
Product concentrations were quantified using a calibration curve
comprised of at least five samples of known concentration. An internal
standard (biphenyl) was used to adjust for variations in the sample
volume.
Electrochemical Methods. Electrochemical measurements were
recorded with a Gamry G300 potentiostat using an electrochemical cell
comprising a 1.5 mm diameter platinum disk electrode, a silver wire
reference electrode, and a platinum wire auxiliary electrode. Cyclic
voltammograms were recorded in a glovebox under an N2 atmosphere at
room temperature in CH2Cl2 solutions containing 1 mM analyte and
0.1 M [Bu4N][PF6] as the supporting electrolyte. All potentials were
referenced to [Cp2Fe]+/0 using an internal [Cp2Co]+/0 standard (E°0 =
ꢀ1.336 V vs [Cp2Fe]+/0).76 Under our conditions, the typical solvent
window is from a reductive limit of ꢀ1.20 V to an oxidative limit of
+1.77 V.
Ligand sterics also play an important role in controlling the
coordination geometry at the metal center of these imido
complexes. The differences in reactivity observed for the reaction
of aryl azides with five-coordinate [NNNcat]TaCl2 and six-
coordinate [NNNcat]TaCl2(py) show that a vacant site on the
metal is critical to azide activation and formation of the metal
imide. The transfer of nitrene to [NNNcat]TaCl2 forms a six-
coordinate imido complex, [NNNq]TaCl2(dNAr), which can
be reduced by 1,2-diphenylhydrazine but is unreactive with
substrates that need to bind first to the metal center. Thus, to
realize nitrene-transfer catalysis, there must be one open co-
ordination site on the metal to accept nitrene from the donor and
a second open coordination site on the metal imido species to
bind the substrate that ultimately will accept the nitrene
fragment. The chemistry of the [NNN]Zr system illustrates
the pitfalls associated with these requirements. [NNNcat]-
ZrCl(THF)2 opens one coordination site to react with ArN3
and form putative [NNNq]ZrCl(dNAr)(THF). The loss of the
second THF molecule from this species generates a second open
coordination site; however, dimerization of two [NNNq]-
ZrCl(dNAr) fragments affords an unreactive imido-bridged
dimer. To effect nitrene transfer, substrate coordination must
outpace dimerization of the [NNNq]ZrCl(dNAr) fragments.
One way to accomplish this result is by using strong donor
ligands such as isocyanides. These ligands both intercept the
terminal zirconium imido species before dimerization and accept
the nitrene fragment to make carbodiimides. Broadening the
nitreneꢀacceptor substrate scope to target products such as
aziridines and amines will require a strategy that slows down
(or stops) the imide dimerization reaction, which, in turn, will
demand redox-active ligands with more sterically demanding
substituents.
Crystallographic Methods. X-ray diffraction data for
[ONOcat]TaCl2(OEt2) 2Et2O, [NN(Me)N]TaMe(dN-p-tolyl), and
3
[ONOcat]Ta(OtBu)2 C5H12 were collected on single crystals mounted
3
on glass fibers with Paratone oil using a Bruker platform diffractometer
equipped with a CCD detector. Measurements were carried out at 143 K
using Mo KR radiation (λ = 0.710 73 Å), which was wavelength-selected
with a single-crystal graphite monochromator. The SMART program
package was used to determine the unit-cell parameters and for data
collection. The raw frame data were processed using SAINT and
SADABS to yield the reflection data file. Subsequent calculations were
carried out using the SHELXTL program. Analytical scattering factors
for neutral atoms were used throughout the analyses. Hydrogen atoms
were generated in calculated positions and refined using a riding model.
ORTEP diagrams were generated using ORTEP-3 for Windows.77 Diffraction
data are shown in Table 2.
Reduction of [NNNq]TaCl2(dNAr) with 1,2-Diphenyl-
hydrazine. Procedure A. An NMR tube with a resealable Teflon valve
was charged with a 0.05 M solution of [NNNq]TaCl2(N-4-C6H4tBu) in
C6D6 (200 μL, 0.01 mmol, 1 equiv), which was diluted with an
additional 300 μL of C6D6. Next, a 0.5 M solution of 1,2-diphenylhy-
drazine (200 μL, 0.1 mmol, 10 equiv) was added using a microsyringe.
The NMR tube was sealed, and the solution was maintained at 35 °C for
24 h. 1H NMR spectroscopy was used to confirm the quantitative
formation of azobenzene and p-tert-butylaniline.
Procedure B. A 1 cm quartz cuvette was charged with 2.5 mL of a
0.18 mM solution of [NNNq]TaCl2(dN-p-tolyl) in THF (0.45 μmol, 1
equiv) and 0.5 mL of a 0.14 M solution of 1,2-diphenylhydrazine in THF
(70 μmol, 155 equiv). The cuvette was sealed, and spectra were recorded
every 30 min at 35 °C for 12 h.
Synthesis of [NN(Me)N]TaMe(dN-p-tolyl). A 20 mL scintilla-
tion vial was charged with a stirbar, 100 mg of [NNNq]TaCl2(dN-p-
tolyl) (0.14 mmol, 1 equiv), and 10 mL of toluene. The solution was
frozen in a cold well. As the solution thawed, 183 μL of MeLi in Et2O
(1.56 M, 0.28 mmol, 2 equiv) was added dropwise. The brown solution
rapidly lightened in color to orange red with concomitant precipitation
VIII. EXPERIMENTAL SECTION
General Synthetic Methods. All compounds reported below are
air- and moisture-sensitive, and as such all manipulations were carried
out under air- and moisture-free conditions using standard Schlenk and
glovebox techniques. Solvents were sparged with argon and deoxygen-
ated and dried by passage through Q5 and activated alumina columns,
respectively. To test for effective oxygen and water removal, solvents
were treated with a few drops of a purple solution of a sodium
benzophenone ketyl radical in THF. Aryl azides,71,72 tertiary alkyl
azides,73,74 [ONOcat]TaCl2(L),46 and [NNNq]TaCl2(dN-p-tolyl)42
were synthesized by literature procedures.
Spectroscopic and Analytical Methods. NMR spectra were
collected on Bruker Avance 600 or 500 MHz spectrometers in dry,
degassed benzene-d6 at 293 K unless otherwise noted. 1H NMR spectra
were referenced to tetramethylsilane using residual proteo impurities of
the solvent (7.16 ppm); 13C NMR spectra were referenced to tetra-
methylsilane using the natural abundance 13C impurities of the solvent
(128.06 ppm).75 All chemical shifts are reported using the standard
9860
dx.doi.org/10.1021/ic200911b |Inorg. Chem. 2011, 50, 9849–9863

Inorganic Chemistry
FORUM ARTICLE
Table 2. X-ray Diffraction Data Collection and Refinement Parameters
[ONOcat]TaCl2(OEt2) 2Et2O
[NN(Me)N]TaMe(dN-p-tolyl)
[ONOcat]Ta(OtBu)2 C5H12
3
3
empirical formula
fw
C36H60Cl2NO4Ta
822.70
C29H39N4O2Ta
656.59
C41H70NO4Ta
821.93
cryst syst
space group
a (Å)
orthorhombic
P212121
14.5421(14)
16.0857(16)
17.0730(17)
90
monoclininc
P21/n
monoclinic
P21/c
14.9724(13)
9.6086(9)
19.6981(17)
90
9.9363(11)
20.418(2)
20.470(2)
90
b (Å)
c (Å)
R (deg)
β (deg)
γ (deg)
volume (Å3)
Z
90
102.8300(10)
90
92.0886(15)
90
90
3993.7(7)
4
2763.1(4)
4
4150.2(8)
4
F(000)
1688
1320
1712
reflns collected
34 532
31 124
9852
indep reflns (Rint
)
9840 (0.0541)
1.101
6732 (0.0186)
1.037
9852 (see CIF)
1.077
GOF
R1 [I > 2σ(I)]a
wR2 (all data)a
0.0582
0.0181
0.0353
0.1514
2
0.0473
0.0774
a R1 = ∑ Fo| ꢀ |Fc /∑|Fo|; wR2 = [∑[w(Fo ꢀ Fc2)2]/∑[w(Fo ) ]]1/2; GOF = [∑w(|Fo| ꢀ |Fc|)2/(n ꢀ m)]1/2
.
2 2
of LiCl. The solution was allowed to warm to ambient temprature over
2 h before it was filtered. Solvent removal gave a yellow solid that was
triturated with 3 ꢁ 5 mL of pentane to give the product in 77% yield
(0.073 g). X-ray-quality crystals were grown from saturated solutions of
the complex in Et2O chilled to ꢀ35 °C. Anal. Calcd for C29H39N4O2Ta:
C, 53.05; H, 5.99; N, 8.53. Found: C, 52.80; H, 5.66; N, 8.40. 1H NMR
[ONOcat]Ta(OtBu)2 as a yellow, crystalline solid (yield 0.052 g, 41%).
X-ray-quality crystals were grown from saturated solutions of pentane
chilled to ꢀ35 °C for 3 days. Anal. Calcd for C36H58NO4Ta: C, 57.67;
1
H, 7.80; N, 1.87. Found: C, 57.36; H, 7.75; N, 1.74. H NMR (600
MHz): δ 7.94 (s, 2H, aryl-H), 1.68 (s, 18H, tBu), 1.45 (s, 18H, tBu), 1.18
(s, 18H, tBu). 13C{1H} NMR (128 MHz): δ154.5 (aryl-C), 146.1 (aryl-
C), 143.5 (aryl-C), 134.4 (aryl-C), 115 (aryl-C), 110.5 (aryl-C), 84.4
(OC(CH3)3), 35.0 (C(CH3)3), 34.8 (C(CH3)3), 32.2 (C(CH3)3), 31.6
(C(CH3)3), 30.1 (C(CH3)3). UVꢀvis (C6H6) λmax/nm (ε/
(600 MHz): δ 7.37 (d, 2H, 3JHH = 6.5 Hz, aryl-H), 6.34 (d, 2H, 4JHH
2 Hz, aryl-H), 7.37 (dd, 2H, 3JHH = 5 Hz, aryl-H), 3.88 (sept, 2H, 3JHH
=
=
5 Hz, CH(CH3)2), 3.43 (s, 6H, OCH3), 3.03 (s, 3H, NCH3), 1.48 (d,
12H, 3JHH = 5 Hz, CH(CH3)2), 1.40 (d, 12H, 3JHH = 5 Hz, CH(CH3)2),
0.83 (s, 3H, TaCH3). 13C{1H} NMR (128 MHz): δ 160.5 (aryl-C),
150.3 (aryl-C), 130.9 (aryl-C), 128.9 (aryl-C), 126.2 (aryl-C), 126.2
(aryl-C), 122.4 (aryl-C), 100.6 (aryl-C), 102.2 (aryl-C), 54.5 (OCH3),
54.0 (CH(CH3)2), 46.9 (TaCH3), 42.7 (NCH3), 25.6 (CH(CH3)2),
24.9 (CH(CH3)2).
M
ꢀ1 cmꢀ1): 290 (15 000).
General Procedure for the Reaction of [ONOcat]TaX2
Complexes with Aryl Azides. In a typical experiment, an NMR
tube with a resealable Teflon valve was charged with 130 μL of a 0.10 M
solution of [ONOcat]TaCl2(OEt2) in C6D6 (0.01 mmol, 1 equiv) and
310 μL of a 0.10 M solution of a 0.44 M solution of PhN3 in C6D6 (0.1
mmol 10 equiv) and diluted with an additional 270 μL of C6D6 using a
microsyringe. The NMR tube was sealed and heated to 55 °C in a
thermostatted water bath. The reaction progress was monitored by 1H
NMR spectroscopy by observing consumption of PhN3 and production
of PhNdNPh. Final quantitation of the azobenzene product was
confirmed by GC/MS.
Titration of [ONOcat]TaCl2(OEt2) with Azobenzene. A stock
solution was prepared by dissolving 37 mg of [ONOcat]TaCl2(OEt2)
(0.05 mmol) in 25 mL of benzene. A second stock solution was prepared
by dissolving 66 mg of (p-tolyl)NdN(p-tolyl) (0.31 mmol) in 25 mL of
benzene. Titration samples for spectroscopic analysis were prepared by
combining 0.5 mL of the [ONOcat]TaCl2(OEt2) stock solution and
various quantities of the diazene stock solution and then diluting to a
total volume of 5 mL with benzene ([Ta]total = 0.2 mM). The samples
were stored at room temperature for several hours to ensure equilibra-
tion before being analyzed by UVꢀvis spectroscopy in 1 cm quartz
cuvettes.
Reaction of [ONOcat]Ta(OtBu)2 with (p-tolyl)N3 with Ex-
cess 1,2-Diphenylhydrazine. An NMR tube with a resealable
Teflon valve was charged with solid [ONOcat]Ta(OtBu)2 (0.013 g, 0.02
mmol, 1 equiv) and 590 μL of a 0.26 M solution of 1,2-phenylhydrazine in
C6D6 (0.2 mmol, 10 equiv). After [ONOcat]Ta(OtBu)2 dissolved,
(p-tolyl)N3 (2 μL, 0.02 mmol, 1 equiv) was added and the NMR tube
was sealed and heated to 55 °C for 48 h. 1H NMR spectroscopy was used to
confirm the formation of 1 equiv of p-toluidine and 1 equiv of PhNNPh.
Synthesis of [ONOcat]Ta(OtBu)2. A 25 mL round-bottomed flask
was charged with a stirbar and a red diethyl ether (15 mL) solution
containing [ONOcat]TaCl2(OEt2) (0.141 g, 0.19 mmol, 1 equiv). The
solution was frozen in a cold well, and as the solution slowly thawed, a
1.44 M solution of MeLi (262 μL, 0.38 mmol, 2 equiv) in Et2O was
slowly dripped in while the solution was stirred. After stirring for 3 h, the
bright-yellow solution was lowered into a cold well and frozen. As the
solution slowly thawed, tBuOH (0.036 mL, 0.38 mmol, 2 equiv) was
slowly added dropwise while the solution was stirred. After stirring for
12 h at ambient temperature, the mixture was filtered and the volatiles
were removed under reduced pressure. The resulting yellow residue was
dissolved in 1 mL of pentane, filtered, and cooled (ꢀ35 °C) to afford
’ ASSOCIATED CONTENT
S
Supporting Information. X-ray crystallographic data in CIF
b
format for [ONOcat]TaCl2(OEt2) 2Et2O, [NN(Me)N]TaMe
3
(dN-p-tolyl), and [ONOcat]Ta(OtBu)2 C5H12. This material is
3
available free of charge via the Internet at http://pubs.acs.org.
9861
dx.doi.org/10.1021/ic200911b |Inorg. Chem. 2011, 50, 9849–9863

Inorganic Chemistry
FORUM ARTICLE
’ AUTHOR INFORMATION
(32) Haneline, M. R.; Heyduk, A. F. J. Am. Chem. Soc. 2006,
128, 8410–8411.
(33) Borgias, B. A.; Cooper, S. R.; Koh, Y. B.; Raymond, K. N. Inorg.
Chem. 1984, 23, 1009–1016.
Corresponding Author
*E-mail: aheyduk@uci.edu.
(34) Blackmore, K. J.; Sly, M. B.; Haneline, M. R.; Ziller, J. W.;
Heyduk, A. F. Inorg. Chem. 2008, 47, 10522–10532.
(35) Blackmore, K. J.; Lal, N.; Ziller, J. W.; Heyduk, A. F. Eur. J. Inorg.
Chem. 2009, 6, 735–743.
’ ACKNOWLEDGMENT
(36) Blackmore, K. J.; Lal, N.; Ziller, J. W.; Heyduk, A. F. J. Am.
Chem. Soc. 2008, 130, 2728–2729.
(37) van der Vlugt, J. I.; Reek, J. N. Angew. Chem., Int. Ed. 2009,
48, 8832–8846.
(38) Gibson, V. C.; Redshaw, C.; Solan, G. A. Chem. Rev. 2007,
The authors thank Dr. Joe Ziller for assistance with X-ray
crystallography, Dr. John Greaves for assistance with mass
spectrometry, and David Lacy for assistance with EPR spectros-
copy. The NSF-CAREER program supported this work (Grant
CHE-0645685). A.F.H. is a Camille Dreyfus TeacherꢀScholar.
107, 1745–1776.
(39) Girgis, A. Y.; Balch, A. L. Inorg. Chem. 1975, 14, 2724–2727.
(40) Speier, G.; Csihony, J.; Whalen, A. M.; Pierpont, C. G. Inorg.
Chem. 1996, 35, 3519–3524.
’ REFERENCES
(41) Chaudhuri, P.; Hess, M.; Weyhermuller, T.; Wieghardt, K.
Angew. Chem., Int. Ed. 1999, 38, 1095–1098.
(42) Nguyen, A. I.; Blackmore, K. J.; Carter, S. M.; Zarkesh, R. A.;
Heyduk, A. F. J. Am. Chem. Soc. 2009, 131, 3307–3316.
(43) Schrock, R. R.; Lee, J.; Liang, L. C.; Davis, W. M. Inorg. Chim.
Acta 1998, 270, 353–362.
(1) Wigley, D. E. Prog. Inorg. Chem. 1994, 42, 239–482.
(2) Gade, L. H.; Mountford, P. Coord. Chem. Rev. 2001, 216, 65–97.
(3) Duncan, A. P.; Bergman, R. G. Chem. Rec. 2002, 2, 431–445.
(4) Hazari, N.; Mountford, P. Acc. Chem. Res. 2005, 38, 839–849.
(5) Walsh, P. J.; Hollander, F. J.; Bergman, R. G. Organometallics
1993, 12, 3705–3723.
(44) Sharma, S. K.; May, P. S.; Jones, M. B.; Lense, S.; Hardcastle,
K. I.; MacBeth, C. E. Chem. Commun. 2011, 47, 1827–1829.
(45) Nguyen, A. I.; Zarkesh, R. A.; Lacy, D. C.; Thorson, M. K.;
Heyduk, A. F. Chem. Sci. 2011, 2, 166–169.
(46) Zarkesh, R. A.; Ziller, J. W.; Heyduk, A. F. Angew. Chem., Int. Ed.
2008, 47, 4715–4718.
(47) Guo, Z.;Swenson, D. C.;Guram, A. S.;Jordan, R. F.Organometallics
1994, 13, 766–773.
(48) Gꢀomez, M.; Gꢀomez-Sal, P.; Nicolꢀas, M. P.; Royo, P. J. Organo-
met. Chem. 1995, 491, 121–125.
(49) Brackemeyer, T.; Erker, G.; Fr€ohlich, R. Organometallics 1997,
16, 531–536.
(50) Strauch, H. C.; Wibbeling, B.; Fr€ohlich, R.; Erker, G.; Jacobsen,
H.; Berke, H. Organometallics 1999, 18, 3802–3812.
(51) Noh, W.; Girolami, G. S. Inorg. Chem. 2008, 47, 535–542.
(52) Labauze, G.; Samuel, E.; Livage, J. Inorg. Chem. 1980,
19, 1384–1386.
(53) Masui, H.; Lever, A. B. P.; Auburn, P. R. Inorg. Chem. 1991,
30, 2402–2410.
(54) Murphy, E. F.; Murugavel, R.; Roesky, H. W. Chem. Rev. 1997,
97, 3425–3468.
(55) Uhrhammer, R.; Su, Y. X.; Swenson, D. C.; Jordan, R. F. Inorg.
Chem. 1994, 33, 4398–4402.
(6) Walsh, P. J.; Hollander, F. J.; Bergman, R. G. J. Am. Chem. Soc. 1988,
110, 8729–8731.
(7) Schaller, C. P.; Cummins, C. C.; Wolczanski, P. T. J. Am. Chem.
Soc. 1996, 118, 591–611.
(8) Bennett, J. L.; Wolczanski, P. T. J. Am. Chem. Soc. 1997,
119, 10696–10719.
(9) Cundari, T. R.; Klinckman, T. R.; Wolczanski, P. T. J. Am. Chem.
Soc. 2002, 124, 1481–1487.
(10) Schaller, C. P.; Wolczanski, P. T. Inorg. Chem. 1993, 32, 131–144.
(11) Blake, R. E., Jr.; Antonelli, D. M.; Henling, L. M.; Schaefer,
W. P.; Hardcastle, K. I.; Bercaw, J. E. Organometallics 1998, 17, 718–725.
(12) Pugh, S. M.; Tr€osch, D. J. M.; Skinner, M. E. G.; Gade, L. H.;
Mountford, P. Organometallics 2001, 20, 3531–3542.
(13) Gomez, M. Eur. J. Inorg. Chem. 2003, 2003, 3681–3697.
(14) Anderson, L. L.; Schmidt, J. A.; Arnold, J.; Bergman, R. G.
Organometallics 2006, 25, 3394–3406.
(15) Kilgore, U. J.; Yang, X.; Tomaszewski, J.; Huffman, J. C.; Mindiola,
D. J. Inorg. Chem. 2006, 45, 10712–10721.
(16) Fryzuk, M. D. Acc. Chem. Res. 2009, 42, 127–133.
(17) Thomson, R. K.; Bexrud, J. A.; Schafer, L. L. Organometallics
2006, 25, 4069–4071.
(18) Silvia, J. S.; Cummins, C. C. J. Am. Chem. Soc. 2009, 131, 446–447.
(19) Tran, B. L.; Pink, M.; Gao, X.; Park, H.; Mindiola, D. J. J. Am.
Chem. Soc. 2010, 132, 1458–1459.
(20) Pierpont, C. G.; Lange, C. W. Prog. Inorg. Chem. 1994, 41,
331–442.
(21) Hendrickson, D. N.; Pierpont, C. G. Top. Curr. Chem. 2004,
234, 63–96.
(56) Jordan, R. F.; Dasher, W. E.; Echols, S. F. J. Am. Chem. Soc.
1986, 108, 1718–1719.
(57) Baldwin, T. C.; Huber, S. R.; Bruck, M. A.; Wigley, D. E. Inorg.
Chem. 1993, 32, 5682–5686.
(58) Burckhardt, U.; Casty, G. L.; Gavenonis, J.; Tilley, T. D.
Organometallics 2002, 21, 3108–3122.
(59) Schmidt, J. A. R.; Arnold, J. Organometallics 2002, 21, 3426–3433.
(60) Bhattacharya, S.; Gupta, P.; Basuli, F.; Pierpont, C. G. Inorg.
Chem. 2002, 41, 5810–5816.
(22) Lever, A. B. P.; Gorelsky, S. I. Struct. Bonding (Berlin) 2004,
107, 77–114.
(23) Poddel’sky, A. I.; Cherkasov, V. K.; Abakumov, G. A. Coord.
Chem. Rev. 2009, 253, 291–324.
(61) Mayer, J. M. Acc. Chem. Res. 2011, 44, 36–46.
(62) Warren, J. J.; Tronic, T. A.; Mayer, J. M. Chem. Rev. 2010,
110, 6961–7001.
(63) Camacho-Camacho, C.; Mijangos, E.; Castillo-Ramos, M. E.;
Esparza-Ruiz, A.; Vasquez-Badillo, A.; Noeth, H.; Flores-Parra, A.;
Contreras, R. J. Organomet. Chem. 2010, 695, 833–840.
(64) Hanna, T. E.; Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc.
2006, 128, 6018–6019.
(65) Dunn, S. C.; Hazari, N.; Cowley, A. R.; Green, J. C.; Mountford,
P. Organometallics 2006, 25, 1755–1770.
(66) Hill, J. E.; Profilet, R. D.; Fanwick, P. E.; Rothwell, I. P. Angew.
Chem., Int. Ed. 1990, 29, 664–665.
(24) Lever, A. B. P. Coord. Chem. Rev. 2010, 254, 1397–1405.
(25) Chaudhuri, P.; Verani, C. N.; Bill, E.; Bothe, E.; Weyherm€uller,
T.; Wieghardt, K. J. Am. Chem. Soc. 2001, 123, 2213–2223.
(26) Herebian, D.; Bothe, E.; Bill, E.; Weyherm€uller, T.; Wieghardt,
K. J. Am. Chem. Soc. 2001, 123, 10012–10023.
(27) Pierpont, C. G. Coord. Chem. Rev. 2001, 216, 99–125.
(28) Evangelio, E.; Ruiz-Molina, D. Eur. J. Inorg. Chem. 2005,
15, 2957–2971.
(29) Chirik, P. J.; Wieghardt, K. Science 2010, 327, 794–795.
(30) Ketterer, N. A.; Fan, H.; Blackmore, K. J.; Yang, X.; Ziller, J. W.;
Baik, M. H.; Heyduk, A. F. J. Am. Chem. Soc. 2008, 130, 4364–4374.
(31) Blackmore, K. J.; Ziller, J. W.; Heyduk, A. F. Inorg. Chem. 2005,
44, 5559–5561.
(67) Evans, D. F. J. Chem. Soc. 1959, 2003–2005.
9862
dx.doi.org/10.1021/ic200911b |Inorg. Chem. 2011, 50, 9849–9863

Inorganic Chemistry
FORUM ARTICLE
(68) MacCarthy, P. Anal. Chem. 1978, 50, 2165–2165.
(69) Klopman, G.; Doddapaneni, N. J. Phys. Chem. 1974, 78,
1825–1828.
(70) Proulx, G.; Bergman, R. G. Organometallics 1996, 15, 684–692.
(71) Smith, P. A. S.; Brown, B. B. J. Am. Chem. Soc. 1951, 73,
2438–2441.
(72) Barral, K.; Moorhouse, A. D.; Moses, J. E. Org. Lett. 2007, 9,
1809–1811.
(73) Miller, J. A. Tetrahedron. Lett. 1975, 16, 2959–2960.
(74) Sasaki, T.; Nakanishi, A.; Ohno, M. J. Org. Chem. 1981, 46,
5445–5447.
(75) Fulmer, G. R.; Miller, A. J. M.; Sherden, N. H.; Gottlieb, H. E.;
Nudelman, A.; Stoltz, B. M.; Bercaw, J. E.; Goldberg, K. I. Organome-
tallics 2010, 29, 2176–2179.
(76) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877–910.
(77) Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565–565.
9863
dx.doi.org/10.1021/ic200911b |Inorg. Chem. 2011, 50, 9849–9863